paper-with-me

홈 › Papers

Adversarial Attack against Cross-lingual Knowledge Graph Alignment

2021-11-01 · EMNLP 2021 11 · Zeru Zhang, Zijie Zhang, Yang Zhou, Lingfei Wu, Sixing Wu, Xiaoying Han, Dejing Dou, Tianshi Che, Da Yan

Recent literatures have shown that knowledge graph (KG) learning models are highly vulnerable to adversarial attacks. However, there is still a paucity of vulnerability analyses of cross-lingual entity alignment under adversarial attacks. This paper proposes an adversarial attack model with two novel attack techniques to perturb the KG structure and degrade the quality of deep cross-lingual entity alignment. First, an entity density maximization method is employed to hide the attacked entities in dense regions in two KGs, such that the derived perturbations are unnoticeable. Second, an attack signal amplification method is developed to reduce the gradient vanishing issues in the process of adversarial attacks for further improving the attack effectiveness.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Tasks

Adversarial AttackEntity Alignment

Similar Papers 제목 키워드 기반

A Generative Adversarial Attack for Multilingual Text Classifiers

2024-01-16 · Tom Roth, Inigo Jauregi Unanue, Alsharif Abuadbba, Massimo Piccardi

Current adversarial attack algorithms, where an adversary changes a text to fool a victim model, have been repeatedly shown to be effective against text classifiers. These attacks, however, generally assume that the vict…

Adversarial Attack

Lost In Translation: Generating Adversarial Examples Robust to Round-Trip Translation

2023-07-24 · Neel Bhandari, Pin-Yu Chen

Language Models today provide a high accuracy across a large number of downstream tasks. However, they remain susceptible to adversarial attacks, particularly against those where the adversarial examples maintain conside…

Machine TranslationTranslation

Towards Understanding the Fragility of Multilingual LLMs against Fine-Tuning Attacks

2024-10-23 · Samuele Poppi, Zheng-Xin Yong, Yifei He, Bobbie Chern 외

Recent advancements in Large Language Models (LLMs) have sparked widespread concerns about their safety. Recent work demonstrates that safety alignment of LLMs can be easily removed by fine-tuning with a few adversariall…

Instruction FollowingSafety Alignment

Code-Mixing on Sesame Street: Dawn of the Adversarial Polyglots

2021-03-17 · NAACL (CALCS) 2021 6 · Samson Tan, Shafiq Joty

Multilingual models have demonstrated impressive cross-lingual transfer performance. However, test sets like XNLI are monolingual at the example level. In multilingual communities, it is common for polyglots to code-mix …

Cross-Lingual TransferXLM-R

BadLingual: A Novel Lingual-Backdoor Attack against Large Language Models

2025-05-06 · Zihan Wang, Hongwei Li, Rui Zhang, Wenbo Jiang 외

In this paper, we present a new form of backdoor attack against Large Language Models (LLMs): lingual-backdoor attacks. The key novelty of lingual-backdoor attacks is that the language itself serves as the trigger to hij…

Backdoor Attack